> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.meetstream.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.meetstream.ai/_mcp/server.

# Claude Integration

The MeetStream Claude Code Plugin brings MeetStream guidance into your development workflow. It helps Claude understand common integration patterns, generate relevant API calls, and guide you through building and debugging meeting bots without repeatedly supplying the same product context.

<Info>
  The plugin is designed for **Claude Code**. Install it from within an active Claude Code session using the commands below.
</Info>

## What the plugin helps you do

<CardGroup cols={2}>
  <Card title="Build meeting bots" icon="robot">
    Create bot workflows for Google Meet, Microsoft Teams, and Zoom from a clear description of your use case.
  </Card>
  <Card title="Configure transcription" icon="closed-captioning">
    Set up live transcription, audio streaming, recording, and speaker-aware meeting data.
  </Card>
  <Card title="Handle events" icon="webhook">
    Generate webhook handlers and callback logic for bot status, transcripts, recordings, and meeting events.
  </Card>
  <Card title="Connect calendars" icon="calendar">
    Work through Google Calendar integrations and the scheduling patterns that support automated bot deployment.
  </Card>
</CardGroup>

## Install the plugin

Open Claude Code in your project, then complete these steps in the Claude Code prompt.

<Steps>
  <Step title="Add the MeetStream marketplace">
    ```text
    /plugin marketplace add meetstream-ai/claude-plugin
    ```

    This registers the MeetStream marketplace and makes its plugins available to Claude Code. It does not install the plugin yet.
  </Step>

  <Step title="Install the MeetStream plugin">
    ```text
    /plugin install meetstream@meetstream-ai
    ```

    Claude Code downloads and installs the MeetStream plugin from the marketplace you added.
  </Step>

  <Step title="Load the plugin">
    ```text
    /reload-plugins
    ```

    Reloading activates the plugin in your current session so its MeetStream knowledge is immediately available.
  </Step>
</Steps>

<Tip>
  These are Claude Code slash commands. Enter them in the Claude Code prompt, not in your operating system shell.
</Tip>

## Start building with Claude

After installation, describe the result you want in natural language. Claude can use the plugin's MeetStream knowledge to plan the integration, generate code, explain configuration, or help diagnose a problem.

For better results, include:

- The meeting platform you are targeting
- The language and framework used by your application
- Whether you need recording, transcription, live media, or chat
- How you want to receive results, such as webhooks or live streams
- Any existing code, errors, or operational constraints Claude should account for

### Example prompts

<Tabs>
  <Tab title="Create a bot">
    ```text
    Add a MeetStream bot to my Node.js application. It should join a Google Meet call, record the meeting, and send transcript updates to my webhook.
    ```
  </Tab>
  <Tab title="Add live transcription">
    ```text
    Show me how to stream speaker-aware transcripts from a MeetStream bot into my Python application and handle reconnects safely.
    ```
  </Tab>
  <Tab title="Debug an integration">
    ```text
    Review this bot creation request and webhook handler. Explain why completion events are not being processed, then propose the smallest safe fix.
    ```
  </Tab>
</Tabs>

## What's included

The plugin provides Claude with guidance for the most common MeetStream development workflows:

- Creating, monitoring, and removing meeting bots
- Configuring recording, transcription, and live audio
- Controlling bots through WebSocket commands
- Processing webhooks and callback events
- Integrating Google Calendar workflows
- Generating practical MeetStream API requests and application code

<Note>
  Treat generated code as part of your normal development workflow: review it, keep credentials in environment variables or a secrets manager, and test meeting behavior before deploying to production.
</Note>

## Connect MeetStream through MCP

<Card title="Set up the MeetStream MCP Server" href="https://docs.meetstream.ai/build-with-ai/meetstream-mcp-server" icon="plug">
  Connect an AI assistant directly to MeetStream tools so it can create bots, retrieve transcripts, inspect meetings, and perform other API actions from a conversation.
</Card>